Geodiversity and biodiversity are fundamental components of natural values of every country. With the increased negative anthropogenic influence on nature, its richness and diversity are becoming increasingly vulnerable and there is a rising need to preserve it for future generations . In recent years, visitors’ interest in non-living natural resources, such as geoheritage sites, has increased worldwide . This variety of natural resources is defined by Gray  as “the range of soil, geomorphological and geological features”. The components of geodiversity that have scientific, educational and aesthetic significance are identified as geoheritage  and the importance of its conservation has been emphasized by many authors [3, 5, 6]. The term geotourism is linked to visits to geoheritage sites and generally to geodiversity [7–9], but only as a separately specialized type of tourism, with geosites in its focus [6, 10]. Therefore, geoheritage has become an important part of tourism offer in many regions and countries, especially those which have not yet solidified their positions in the tourism market. Visitors are becoming more sophisticated in selecting geotourism destinations they want to visit .
In the City of Belgrade, as well as in the rest of Serbia, there are resources which have not been adequately exploited for the development of geotourism [12–14] and therefore, in global terms, the city and the country represent a very small tourist market [15, 16]. Given that the importance of geodiversity on the territory of Belgrade has completely been minimized and neglected in relation to biodiversity, the aim of this study is to carry out a quantitative and qualitative analysis of geoheritage sites and assessment of their values to promote the development of geotourism as a complementary part of tourism offer in Belgrade. It is important to note that geoheritage sites, as specially adapted locations, may contribute to sustainable development of the area both in economic and environmental terms and that as such, they can successfully be positioned in the tourism market .
Geoheritage sites in various European countries used to be placed under protection based on different criteria, which is why the European Association for the Conservation of the Geological Heritage - Pro-GEO was established in 1995 [17, 18]. The first joint task was to make a European list of geoheritage. All member states have been divided into working groups at the regional level. Serbia belongs to the Pro-GEO working group for Southeastern Europe, the so-called Pro-GEO WG-1. In 1995, the National Council for Serbian Geoheritage was formed, which then established a unique policy of systematized conservation of geoheritage sites. In 1996, the National Council launched a project called the Inventory of Serbian Geoheritage Sites. Its purpose was to select important geoheritage sites for public attention and for conservation against devastation. So far, 651 geoheritage sites have been identified, while 80 of them have already been protected . Current trends in the study and protection of geodiversity elements of an area, education of the population and their presentation to the public as an integral part of the tourist offer show a need for carrying out new research into Belgrade geoheritage sites, as a part of Serbian geoheritage. The literature on selection, registration, promotion and evaluation of Belgrade geoheritage is still quite scarce. The geological monuments located in the territory of Belgrade were first protected in 1968. However, the public and visitors have expressed a growing interest in geoheritage sites in Belgrade over the last ten years [20, 21]. Researches such as Banjac and Rundić  were at that time focused on the topic of geotourism, while Belij , Mijović and Stefanović  and Marković et al. [24–26] wrote about geodiversity and geoheritage in the city and wider area.
All these studies emphasized the need to identify geoheritage sites located in Belgrade and evaluate them in an adequate manner, as well as present them to the public as part of the relevant natural heritage in the best possible way. For these reasons, papers that deal with systematization, presentation and popularization of these sites are of particular relevance. In this respect, we analyzed and compared the current state and tourism potential of sites in Belgrade by using the combination of the GAM/M-GAM models for assessment of geosites. The aim of the paper was to show the relevance of every sub-indicator for the entire assessment process for visitors by comparing them with experts’ opinions. The main goal of the research was to show, by applying the combination of the GAM/M-GAM models to selected geosites, which sub-indicators most influence visitors’ opinion when giving preference to one geosite over another. Afterwards, we presented the results of the assessment for both segments to see how the difference in importance for each sub-indicator has affected the research results.
2 Study area and description of the assessed geosites
The City of Belgrade lies on the slope between the alluvial plains of the great European rivers, the Danube and the Sava. The mathematical coordinates of the observed area are between 44° 39’ and 44° 49’ north latitude and 20° 17’ and 20° 37’ east longitude. There are 37 protected natural resources in the urban area of the city, but most of them are biodiversity sites. The geological diversity of the terrain, which is composed mainly of sedimentary and magmatic rocks of Jurassic-Cretaceous, Tertiary and Quaternary age , is very rich. Also, not all sites have been explored to the same extent. Geomorphological and hydrological sites are not numerous nor do they differ very much in appearance. Due to this fact, only eight main geoheritage sites have been identified in the observed territory, which are officially listed on the Inventory of Serbian Geoheritage Sites. The sites are marked as G1-G8 (Figure 1 and Figure 2). Although not numerous, these sites can be consolidated into a unique tourist tour, which could include similar sites in other parts of Serbia and neighboring countries.
G1: Straževica profile represents a Lower Cretaceous section with the oldest preserved rock formations on the territory of Belgrade discovered on the Straževica hill nearby Rakovica Monastery. According to Rundić , the site is well-known for its Jurassic or the so-called Straževički limestones. The fossils that have been found in this area belong (in most cases) to brachiopods, but there are also gastropods, corals, crinoids and some algae. They are in contact with the Lower Cretaceous, Aptian marls. The limestones have gradually transformed into clay-marl series of Lower Cretaceous.
G2: Mašin Majdan-Topčider is a 20-m-high section of reef limestones, a Cretaceous rock complex from the Senonian Age made of compact bluish rocks with calcite veins. Some of the beds are up to 1 m wide. There are also lenses of shales, marls and sandstones. The fossils of bivalves, corals, gastropods, foraminifera and pachiodont shells can be found in the rocks. At the top of the section there are layers of marsh loess, a rare phenomenon in loess formation. This profile has been protected since 1969 .
G3: Profile at the Kalemegdan Fortress, located at the foot of the Pobednik (Eng. Victor) monument, shows large sections through Badenian reefs with characteristic fauna and shallow-water, coastal, and reef deposits of the former Pannonian Sea. According to Rundić , first fossil discoveries at this section were made in 1886. Here is the core of the Kalemegdan anticline, which is indicated by a series of layers positioned diagonally between the walls of the fortress. In the lower part of the section there are conglomerates, quartz sands moving into sandy-sandstone deposits. The site was for the first time placed under protection in 1968, as the first protected natural monument in the city.
G4: Abandoned quarry in Barajevo represents a shift from the Lower Sarmatian to the Middle Sarmatian period, where the development of the Middle Sarmatian on the territory of the middle and western part of Serbia can be seen. As Rundić et al.  stated, the section is more than 200 m wide, and is composed of sandy limestones, limestone consisting of shell fragments and limestone with mollusk fauna, with inter-layers of siltstones and shales. There are also foraminifera, ostracods and bryozoans. This is a typical example of merokarst, with dry or flooded sinkholes.
G5: Karagača valley is a globally relevant example of the coastal, clastic development of the Upper Pannonia. It is located on the right bank of the Karagača stream, near Vrčin village, at a length of 10 m and a height of 7 m. It consists of coarse sand grains, microconglomerates with intercalations of sandy clay and yellow sands that lie transgressive across the serpentinites. The proximity of older volcanic rocks and strong hydrothermal activity have had a favorable effect on the water chemistry and encouraged development of rich endemic fauna [28, 29].
G6: Artesian well in Ovča was discovered on the left bank of the Danube in 1939. In 1985, a 162-m-deep exploration and exploitation well was built here. According to Rundić , aquifers with water under pressure are located at a depth of 158 m. The base is composed of Badenian-Sarmatian marls and limestones, with deposits of Pliocene clay over them. The top layer is composed of Quaternary sands and coarse gravel grains.
G7: Kapela loess profile represents the one of the most important European loess sections situated along steep cliffs the right bank of the Danube. The profile, located near Batajnica town, has representative loess and paleopedological sections and forming the steep scarps towards the Danube. The Kapela (lit. Chapel) is the section of the Late and Middle Pleistocene loess and fossil soils about 40 m thick. Within this section, there are also tuff interlayers indicating volcanic activity, which increases the chronostratigraphic value of this section [27, 30].
G8: Lake in Sremcica (formerly known by its old name was Rakina Bara) is located at the bottom of an approximately 300 m long and 150 m wide sinkhole. The emergence of the lake is linked to the processes related to Belgrade merokarst, through the Sarmatian limestone. It is characterized by alluvial sinkholes, small depressions and caves, ponds, hanging valleys and small springs and sources. The dimensions of this site vary depending on the researcher and measurement time. More recent measurements indicate that the lake is 170 m long and 110 m wide .
All these eight geosites will be further evaluated by using GAM/M-GAM methods. The overall assessments will be conducted for each site (G1-G8) respectively.
3 Materials and methods
The methodology is based on the modified version of the geosite assessment model (M-GAM) proposed by Tomić and Božić  and tested by Różycka and Migoń . The M-GAM represents a modification of the original GAM model created by Vujičić et al.  and tested by Petrović et al. . Both versions of the models were employed in this survey in order to compare the results of both respondents’ groups. The GAM/M-GAM model was used through the initial workflow presented in Figure 3.
While the GAM model involves grades given by experts, M-GAM includes not only expert opinions, but also the views of visitors regarding the importance of each indicator in the assessment process. The GAM model contained analysis of two key indicators: the main values (MV) and additional values (AV) of geoheritage sites, comprising a total of 27 sub-indicators (Table 1). The main values have 12 sub-indicators, while additional values have 15 sub-indicators. Their numerical values range from 0 to 1, in the following order: 0.00, 0.25, 0.50, 0.75 and 1.00. Sub-Indicator grades and their explanations are shown in detail in Table 2.
Structure of the (original) GAM model.
|Main values (MV)|
|Scientific/Educational value (VSE)|
|VSE1: Rarity||Number of closest identical sites.|
|VSE2: Representativeness||Didactic and exemplary characteristics of the site resulting from its quality and general configuration.|
|VSE3: Knowledge of scientific issue||Number of papers in geo-scientific publications.|
|VSE4: Interpretation||Interpretive possibilities concerning geological and geomorphologic processes, phenomena and shapes and level of scientific knowledge.|
|Scenic/Aesthetic values (VSA)|
|VSA1: Viewpoints||Number of viewpoints accessible by a pedestrian pathway. Each must present a particular angle of view and be situated less than 1 km from the site.|
|VSA2: Surface area||Entire surface area of the site. Each site is considered in quantitative relation to other sites.|
|VSA3: Surrounding landscape and nature||Panoramic view quality, presence of water and vegetation, absence of human induced deterioration, vicinity of urban area, etc.|
|VSA4: Environmental setting||Contrast to nature, contrast of colors, appearance of shapes, etc.|
|VPr1: Current condition||Current state of the geosite.|
|VPr2: Protection level||Protection by local, regional groups, national government, international organizations, etc.|
|VPr3: Vulnerability||Vulnerability level of the geosite.|
|VPr4: Suitable number of visitors||Proposed number of visitors on the site at the same time, according to the surface area, vulnerability and current state of the geosite.|
|Additional values (AV)|
|Functional values (VFn)|
|VFn1: Accessibility||Possibilities for accessing the site.|
|VFn2: Additional natural values||Number of additional natural values in the in radius of 5 km.|
|VFn3: Add. anthropogenic values||Number of additional anthropogenic values in the in radius of 5 km.|
|VFn4: Vicinity of urban centers||Proximity of urban centers.|
|VFn5: Vicinity of road network||Proximity of important road networks in the in radius of 20 km.|
|VFn6: Additional functional values||Parking lots, gas stations, mechanics, etc.|
|Tourism values (VTr)|
|VTr1: Promotion||Level and number of promotional material.|
|VTr2: Organized visits||Annual number of organized visits to the geosite.|
|VTr3: Vicinity of visitor center||Proximity of the visitor center to the geosite.|
|VTr4: Interpretative panels||Text and graphics, material quality, size, if it fits the surroundings, etc.|
|VTr5: Number of visitors||Annual number of visitors.|
|VTr6: Tourism infrastructure||Pedestrian ways, rest stops, garbage cans, toilets, wellsprings. etc.|
|VTr7: Tour guide service||Expertise level, knowledge of foreign language(s), interpretative skills, etc.|
|VTr8: Hostels||Hostels close to the geosite.|
|VTr9: Restaurants||Restaurants close to the geosite.|
Numerical GAM indicators and their description
|GAM Sub-||Grades (0-1)|
|VSE1||Common||Regional||National||International||Exclusive to the geosite|
|VSE3||None||Local publications||Regional publications||National publications||International publications|
|VSE4||None||Limited example of processes, but hard to explain to non-experts||Good example of processes, but hard to explain to non-experts||Limited example of processes, but easy to explain to the common visitor||Good example of processes and easy to explain to the common visitor|
|VSA1||None||1||2 to 3||4 to 6||More than 6|
|VPrl||Totally damaged (because of human activities)||Greatly damaged (as a result of natural processes)||Medium-level damage (with essential geomorphologic features preserved)||Slightly damaged||No damage|
|VPr3||Irreversible (with possibility of total loss)||High (could be easily damaged)||Medium (could be damaged by natural processes or human activities)||Low (could be damaged only by human activities)||None|
|VPr4||0||0 to 10||10 to 20||20 to 50||More than 50|
|VFnl||Inaccessible||Low (on foot, with special equipment and expert guide tours)||Medium (by bicycle and other means of man-powered transport)||High (by car)||Utmost (by bus)|
|VFn2||None||1||2 to 3||4 to 6||More than 6|
|VFn3||None||1||2 to 3||4 to 6||More than 6|
|VFn4||More than 100km||100 to 50km||50 to 25km||25 to 5km||Less than 5km|
|VTr2||None||Less than 12 per year||12 to 24 per year||24 to 48 per year||More than 48 per year|
|VTr3||More than 50km||50 to 20km||20 to 5km||5 to 1km||Less than 1km|
|VTr4||None||Low quality||Medium quality||High quality||Utmost quality|
|VTr5||None||Low (less than 5000)||Medium (5001 to 10.000)||High (10.001 to 100.000)||Utmost (more than 100.000)|
|VTr8||More than 50 km||25-50 km||10-25 km||5-10 km||Less than 5km|
|VTr9||More than 25 km||10-25 km||10-5 km||1-5 km||Less than 1 km|
The importance factor (1 ≥ Im ≥ 0), where each respondent was asked to rate all presented sub-indicators, was included in the survey. The importance factor can be a very useful examination tool because it gives visitors the chance to express their attitude to every single sub-indicator in the model . Moreover, in the case of Belgrade geosites, the importance factor has emphasized the relevance of visitors having an opportunity to choose which city’s non-cultural attractions they are going to visit and their attitudes thereto [35, 36]. It was necessary to include visitors in the survey, mainly because experts can cover only the marketable aspects of geosites. Moreover, the experts carried out their evaluation from the scientific perspective, which, as research has shown, is usually less important to the average visitor. Nevertheless, expert opinions combined with those of regular visitors provide more objective and accurate results. Visitors can rate the sub-indicators in the same manner as experts rate them for both groups of values (by giving them exact numerical values from 0.00 to 1.00). The main values (MV) consist of the following indicators:
- Scientific and educational values - VSE (rarity, representativeness, knowledge of geo-scientific issues and interpretation)
- Aesthetic values - VSA (viewpoints, surface, surrounding landscape and nature and environmental setting) and
- Protection values - VPr (current condition, protection level, vulnerability and acceptable number of visitors). The MV is calculated as a sum of the presented sub-indicators:
On the other hand, additional values are composed of:
- Functional values - VFN (accessibility, additional natural values, additional anthropogenic values, vicinity of urban centers and important road networks and additional functional values) and
- Tourism values - VTr (promotion, number of organized visits, number of visitors, vicinity of visitor centers, interpretive panels, tourism infrastructure, accommodation, restaurant service and quality tour guide service) [6, 34]. The AV is calculated as:
By adding together MV and AV, we get the following equation:
To reach the M-GAM, it is necessary to add in the importance factor (Im), which is calculated as:
In this equation, Ivk represents the score of visitors’ mark per sub-indicator, K is the total number of visitors, while the Im parameter can have any value in the range from 0.00 to 1.00. Grades given by experts and visitors for each sub-indicator are shown in detail in Table 3. Finally, by adding all the values together, we arrive at the following equation:
Scores given by experts and visitors for every sub-indicator.
|GAM Sub-indicators||Scores by experts 0.00-1.00)||lm Factor||Total scores 0.00-1.00)|
|VSE - Scientific and educational values|
|VSA - Scenic and aesthetic values|
|VPr - Protection values|
|VFn - Functional values|
|VTr - Tourism values|
As it can be seen from the M-GAM equation, the importance factor (Im) is multiplied by the grade given by experts. Therefore, a more realistic assessment is carried out by using the M-GAM. This can be concluded from the fact that if the importance of a sub-indicator is graded 0.50 by visitors, the final mark cannot be 0.75 or 1.00, but instead, it should be lower (0.50) if visitors’ attitudes are also considered. In this respect, the values of M-GAM indicators are constantly equal to or lower than GAM values. The M-GAM model aims to show the status of main and additional tourist values of geoheritage sites which have not yet reached their maximum potential. This gives us a realistic picture of geoheritage, based on which it is possible to plan and promote tourism activities for the analyzed sites . However, sub-indicators with lower grades are not so important for the development of tourism . By using the M-GAM model, the above-mentioned authors have tried to shift the main focus from the opinions of experts to those of visitors regarding the significance of tourism values of geosites. This means that the future development of tourism at the observed Belgrade geoheritage sites should be promoted by improving the values whose potentials have not yet been realized, but which are important to visitors.
A total of 216 visitors and 63 experts filled out the questionnaire properly. The questionnaire conducted in the research consisted of 27 questions/sub-indicators. It consists of two complementary parts. The first part involves items which concerned many values, while the second part includes additional values of the GAM model. Every respondent was asked to rate the importance factor (Im) of every sub-indicator on a five point Likert-type scale by rating it from zero to one (0 = not at all important; 0.25 = not very important; 0.50 = neutral; 0.75 = somewhat important; 1.00 = very important). A survey was conducted among people who visited Belgrade between April and September of 2016, as well as among experts in geotourism and physical geography (geomorphology and geology) from Serbia and abroad. Sampling was convenient since subject were only visitors and experts willing to participate in the study. At the beginning, we informed the respondents about the subject of examination. Within our study, we investigated the information seeking of respondents of both genders, different educational levels, personal incomes, many kinds of occupations, etc.
4 Results and discussion
The research results were obtained using a sum of main and additional values and their mean values. Summary scores, given by experts and visitors for individual sites, are shown in Table 3.
Results for the five main comparison criteria shown in Table 3 demonstrate that the Karagača valley near Vrčin (G5) has the highest scientific and educational value according to the experts’ opinion. The same site has the highest score according to the total value (visitors’ opinion included). This can be justified by the fact that G5 represents a globally relevant example of the coastal, clastic development of the Upper Pannonia in this part of the Pannonian Basin. It has thus received highest scores for this criterion among all observed geosites. Due to the findings recorded at the G5 section, the international scientific community has decided to call the younger stage of development of the Pannonia in the Central Paratethys “Serbian”. Together with these, some representatives of molluscs and ostracods are determined here for the first time and named after this section [27–29].
When considering the scenic and aesthetic values, attitudes of experts, as well as those of visitors are quite the same. Both groups of respondents gave highest scores to the profile at Kalemegdan Fortress (G3). The scores can be explained by the micro-location of the geosite, which is situated at the foot of the Victor monument, a famous Belgrade landmark and one of the most popular scenic symbols of the city. The profile faces the confluence of the Sava and the Danube and has the view of the vast Pannonian plain. Below this landmark, the site shows representative sections of Badenian reefs, with characteristic shallow-water fauna and coastal and reef deposits of the former Pannonian Sea .
As regards protection values, overall scores are significantly distinctive due the fact that the experts gave the highest scores for the profile at the Kalemegdan Fortress (G3), in the same way as visitors marked this site as the most protected one. This profile had a long tradition of protection (since 1968), low level of vulnerability (not threatened significantly by any direct natural or anthropological causes) and is presently in good condition, which was enough for the both groups to give it the highest numerical scores.
When analyzing the importance of sub-indicators in respect of functional values, it can be noticed that the highest scores, as evaluated by all respondents, were also given to the profile at the Kalemegdan Fortress (G3). The experts, as well as the visitors, assessed that the level of the accessibility of this site, its additional attractions, proximity of urban centers and traffic networks was highest. This can easily be explained by the obvious fact that this particular geosite is located in the center of a typical urban setting (the territory of the City of Belgrade has approximately 1.3 million inhabitants according to the national census data ), with high level of urban infrastructural development and numerous cultural and natural attractions in its closest surroundings. Similar applies to other analyzed geosites.
As regards the evaluation of tourism values, the situation is not so different. Apparently, further results have shown that both groups share a positive attitude towards the same geosite located in downtown Belgrade, namely the profile at the Kalemegdan Fortress (G3). In this regard, the site has achieved the highest score of 8.50/7.68. However, results for the total value indicate that its promotion, organized visits to the site and number of visitors on the site are “the weakest points” and that these segments are its main disadvantages. On the other hand, the experts evaluated promotion with lower score (0.50), i.e. they shared the visitors’ opinion that the quantity of promotional material and level of its utilization were not as good as they could be. The position of this geoheritage site on the tourism marked could be improved by more effective collaboration with tourism organizations, as well as by making more effective international offer  that could improve and develop conservation and promote the geosite in a much wider region.
Paragraphs below provide more detailed explanation and present the role of the Im parameter. In further analyses, Tables 4 and 5 and Figure 4 show the findings of the assessment obtained by using both GAM/M-GAM models.
Overall assessments of the observed geosites by using GAM.
Overall assessments of the observed geosites by using M-GAM.
The average of main values was 6.84 and 8.22 for additional values within the GAM model (Table 4). Three sites had the largest sum of main values (n≥ 8), namely the profile at the Kalemegdan Fortress (MV=9.75), the Karagača stream valley near Vrčin (MV=8.25) and the Kapela loess profile near Batajnica (MV=8.25). Moreover, four sites had the largest sum of additional values (n≥ 8), namely the profile at the Kalemegdan Fortress (AV=14.50), Mašin Majdan-Topčider (AV=10.50), the Kapela loess profile near Batajnica, (AV=9.25) and the artesian well in Ovča (AV=8.50). In both cases, the G3 site has scored highest within the GAM model, which points to a conclusion that this site represents the most significant part of geotourism offer in Belgrade. On the other hand, Straževica profile (G1) scored lowest in the case of main values (MV=4.50), while the lake in Sremciča (G8) scored lowest in the case of additional value (AV =3.00). The reasons for these scores are that both experts and visitors who were included in the GAM assessment assessed some sub-indicators as less relevant.
Results from Table 5 indicate that the average of main values within the M-GAM model was MV=5.32, while the average for additional values was AV=7.14. According to the presented data, none of the observed sites had the largest sum of main values (n≥ 8), while only two sites had the largest sum of additional values (n≥ 8), namely the profile at the Kalemegdan Fortress (AV=12.78) and Mašin Majdan-Topčider (AV=9.07). On the other hand, the lowest sum of main values was characteristic of the Straževica section (MV=3.55), while the lake in Sremciča obviously had the lowest sum of additional values (AV=2.70). It can be noticed that the same situation has been noticed in GAM calculations results and the reason is that some sub-indicators were also assessed as less relevant by both groups of respondents involved in the M-GAM.
When comparing the position of the observed geoheritage sites in the GAM/M-GAM matrices (Figure 4), it is obvious that the distinctions in sites’ positions indicates different results of the assessment done by experts exclusively, as well as by visitors and experts jointly. Depending on the determined values obtained by the assessment, every geosite could be put in one of the fields in the matrix, divided into nine zones. The sum of MV and AV scores for every individual geosite is presented via X and Y axes respectively. Both matrices are indicated by Z(i, j) fields where (i, j = 1, 2, 3), according to the grades they have received during the evaluation process. The fields are created by grid lines, which show the exact level of GAM/M-GAM indicators, according to the positions of the MV and AV scores.
The findings presented in Figure 4 indicate that that five of eight assessed geosite has changed its Z(i, j) field position in the M-GAM matrix in comparison to the (primary) GAM matrix. With the exceptions of the abandoned quarry in Barajevo (G4), the artesian well in Ovča (G6) and the lake in Sremciča (G8), every other site has transformed its position. This can be explained by the fact that their AVs were quite lower to begin with, so when multiplied by the importance factor, they did not change significantly. On the other hand, changes in positions can be noticed in cases of the Straževica profile (G1), Mašin Majdan-Topčider (G2), profile at the Kalemegdan Fortress (G3), Karagača valley (G5) and Kapela loess profile (G7). As regards the position of G1, it can be noticed that both MV and AV for this geosite have lower positions and that it has moved from field Z22, representing moderate AV and moderate MV, to field Z12, representing low MV and moderate MV. The position of G2 is similar since both MV and AV are lower in the M-GAM matrix. G2 has moved from field Z23, representing moderate MV and high AV, to field Z22, representing moderate MV and AV. Together with the previous two cases, G3 has also lower MV and AV, because it has moved from field Z33, with the highest values of both MV and AV, to field Z23, representing still high AV, but moderate MV score. Contrary to other presented geosites, G4 position has moved lower, but very slightly and within the same field Z22. Almost the same case involves the position of G6, with very gentle modification (both MV and AV have been decreased for a bit). These results are due to the fact that their MVs were not much lower to begin with, so when multiplied by the importance factor, they have not changed significantly. On the other hand, the position of G5 has changed considerably from field Z32, representing high MV but moderate AV, to field Z22, representing moderate MV and moderate AV. Very similar situation is obvious in the case of G7, where it can be perceived that MV has moved from field Z32, representing high MV, but moderate AV, to field Z22, representing moderate MV and AV. Finally, the position of G8 is almost equal in both matrices (slightly changing position within Z21 representing moderate MV and low AV). In general, all observed sites have moved to lower positions – mainly caused by additional values decline but also because the modifications of main values position. The presented distinctions can be explained by the fact that additional values are generally less important to experts, which has influenced the lower position of certain geosites in their assessment. In terms of main values, we can see that they are almost equally important to both groups of respondents, so they have not influenced the position of geosites in any radical or mayor way.
The combination of the GAM/M-GAM models can yield more precise and objective results of assessment of main and additional values of the observed Belgrade geosites. Moreover, a clearer and more realistic picture is thus obtained, which can be rather useful for planning and improvement of visitors’ activities at other geosites in the country. This is because not all indicators can have the same weight, as presented in the original GAM, since visitors can attach different levels of relevance to different (sub-)indicators, when choosing whether to visit a certain geosite or not. Therefore, this is very relevant issue that must be considered in the overall assessment of geosites.
By summing the final findings for all analyzed Belgrade geosites, we can draw a conclusion that when assessing a geosite, experts appreciated values that were considerably different from those that were relevant to visitors. Consequently, the results that included the attitudes of visitors were markedly different. The scientific and educational values (VSE) for all eight geosites seemed to be important to geosites’ visitors when choosing their destinations. This specifically referred to sub-indicators such as rarity (VSE1), representativeness (VSE2) and knowledge of scientific issues related to geosciences (VSE3) because the difference between results was not so extreme. However, the sub-indicator of interpretation (VSE4) proved to be a not very important factor for them, as we can see from the fact that they gave it a score of 0.68, which has changed the results significantly.
When analyzing the second group of sub-indicators (VSA), we can conclude how the importance factor for some of them, as assessed by visitors, can considerably change the assessment’s results. For example, the viewpoints sub-indicator (VSA1) was rated by visitors as the factor of greatest importance (Im = 1.00), which means that it played a significant role in visitors’ opting for a place to visit. In addition, when marks given by experts are multiplied by the importance factor (Im) rated by visitors, we get similar results, so there is no significant change. Together with this, the surface area sub-indicator (VSA1) got a very high score (Im = 0.92), which means that visitors appreciated greatly the whole micro-surface of the observed site. However, the sub-indicators such as surrounding landscape and nature (VSA3) and the environmental setting of sites (VSA4), which was chosen by experts as an important factor to be included in the assessment, did not seem to be of the same relevance for the visitors (in both cases, Im = 0.57). It can be seen how this has affected the final findings since the grades given by experts were multiplied by the importance factor as assessed by the visitors, which produced lower results.
When it comes to the protection value (VPr), importance values of current condition (VPr1, Im=1.00), vulnerability (VPr3, Im=0.75) and acceptable number of visitors (VPr4, Im=0.92) scored very high. These high scores given by the experts should not be taken as completely realistic for the assessment. The reason is that the visitors attached only minor significance to some of these sub-indicators, which did not have any real effect on their decision to visit a site. This especially refers to the protection level sub-indicator (VPr2), which was highly rated by experts, but the visitors did not rate it as significant at all (Im=0.54), so the final score should be much lower.
Additionally, not all functional values, as it was the case with the main values, were of the same importance for visitors of the observed geosites. Here, once again, we can see how this fact can fundamentally change the final findings. For example, even though there are plenty of additional natural values (VFn2) in the near surroundings of these geosites (as can be seen from the highest grade given by the experts), this sub-indicator did not seem to be less important to visitors (Im=0.74) in comparison with some other sub-indicators, such as the vicinity of important road networks (VFn5, Im=1.00) or the additional functional values (VFn6, Im=0.91), which got the highest scores.
On the other hand, the tourism values (VTr) are (normally) the most important to visitors since relevance for most of the sub-indicators was higher than Im≥ 0.86 (i.e. the highest scores were given to the vicinity of the visitor center, interpretative panels, number of visitors, tourism infrastructure, tour guide service, hostels and restaurants). However, here we can also notice some exceptions, such as in the case of the promotion sub-indicator (VTr1, Im=0.50), as well as the annual number of organized visits (VTr2, Im=0.77). The experts considered these to be important for the overall assessment, which did not match visitors’ opinions. This is one more proof that we cannot rely solely on the opinion of experts, which are just one group of tourists that visit these sites. The exclusion of other walks of life and their opinions can only yield GAM results that are less objective and accurate than those obtained by using its modified version (M-GAM), where other segments of society beside experts are also included in the assessment.
It can thus be concluded, and these findings need to be emphasized, that the perception of Belgrade geosites differs at numerous levels observed and that there are no additional sites that would enable the tourist functioning of an area and a more complex development of tourism. Also, it is necessary to consolidate all natural and anthropogenic motifs from this area into a complex tourism value or incorporate these sites into a unique tourist tour since if they remain unintegrated, they will only have the character of complementary touristic value of the City of Belgrade.
The research was supported by the Ministry of Education, Science and Technological Development, Republic of Serbia (Grant III 47007) and by Tomsk Polytechnic University, Russian Federation (14.Z50.31.0029 from March 19, 2014).
Erikstad L., Geoheritage and geodiversity management – the questions for tomorrow. Proceedings of the Geologists’ Association, 2013, 124, 4, 713–719
Newsome D., Dowling R., Leung Y-F., The nature and management of geotourism: A case study of two established iconic geotourism destinations. Tourism management perspectives, 2012, 2–3, 19–27
Gray, M., Geodiversity. Valuing and conserving abiotic nature. Wiley, Chichester, 2004
Dixon G., Geoconservation: An International Review and Strategy for Tasmania. Miscellaneous Report. Parks and Wildlife Service, Tasmania. 1996, 1-101
Giurginca, A., Munteanu, C.M., Stanomir, M.L., Niculescu, G., Giurginca, M., Assessment of potentially toxic metals concentration in karst areas of the Mehedinti plateau geopark (Romania). Carpathian Journal of Earth and Environmental Sciences, 2010, 5, 1, 103–110
Petrović, M.D., Vasiljević, Dj.A., Vujičić, M. D., Hose, T. A., Marković, S.B., Lukić, T., Global geopark and candidate – comparative analysis of Papuk Mountain geopark (Croatia) and Fruška Gora Mountain (Serbia) by using GAM model. Carpathian Journal of Earth and Environmental Sciences, 2013, 8, 1, 105-116
HoseT.A., Geotourism in England: a two-region case study analysis. Unpublished PhD thesis in 2 volumes, University of Birmingham, UK, 2003
Hose T.A., Geotourism in Almeria Province, southeast Spain. Turizam, 2007, 55, 3, 259-276
Hose T.A., Towards a history of geotourism: definitions, antecedents and the future. In: Burek C.V., Prosser C.D. (Eds.), The history of geoconservation (Special Publication 300). The Geological Society, London, 2008, 37-60
Plyusnina, E. E., Ruban, D. A., Zayats, P. P. Thematic dimension of geological heritage: An evidence from the Western Caucasus. Journal of the Geographical Institute “Jovan Cvijic” SASA, 2015, 65, 1, 59-76
Božić S, Tomić N. Canyons and gorges as potential geotourism destinations in Serbia: comparative analysis from two perspectives – general geotourists’ and pure geotourists’. Open Geosciences, 2015, 7(1), 531-546
Vasiljević Dj.A., Marković S.B., Hose T.A., Smalley I., Basarin B., Lazic L., Jovic G., The introduction to geoconservation of loess palaeosol sequences in the Vojvodina region: Significant geoheritage of Serbia. Quaternary International, 2011a, 240, 108–116
Vasiljević Dj.A., Marković S.B., Hose T.A., Smalley I., O’Hara-Dhand K., Basarin B., Lukić, T., Vujičić, M.D., Loess towards (geo) tourism – proposed application on loess in Vojvodina region (north Serbia). Acta geographica Slovenica, 2011b, 51(3), 391-406
Jojić-Glavonjić, T., Milivojević, M., Panić, M., Protected geoheritage sites as a touristic value of Srem. Journal of the Geographical Institute “Jovan Cvijic” SASA, 2014, 64, 1, 33-50
Hose T.A., Geo-tourism - appreciating the deep time of landscapes. In: Novelli M. (Ed.), Niche Tourism: contemporary issues, trends and cases. Elsevier Science, Oxford, 2005, 27–37
Novelli, M., Benson A., Niche tourism: A way forward to sustainability? In: Novelli M. (Ed.), Niche Tourism: contemporary issues, trends and cases. Elsevier Science, Oxford, 2005, 247–251
Wimbledon, W.A.P., Ishchenko, A.A., Gerasimenko, N.P., Karis, L. O., Suominen, V., Johansson, C.E. Freden, C., Geosites - an IUGS initiative: science supported by conservation. In: Barretino, D, Wimbledon, WP, Gallego E (Eds.), Proceedings of the Geological heritage: its conservation and management, Madrid, Spain. Instituto Tecnologico Geominero de Espana, Madrid, 2000, 69-94
Joksimović, M.M., Gajić, M.R., Vujadinović, S.M., Golić, R.M., Vuković, D.B., The effect of the thermal component change on regional climate indices in Serbia. Thermal Science, 2015, 19(2), 391-403
Mijović D., Stefanović, I., Inventar objekata geonasleđa Srbijeod ideje do optimalnog modela (The inventory of Serbian geoheritage site -from idea to optimal model). Protection of Nature, 2008, 60, 1-2, 359-365 (in Serbian with English summery)
Grubacčvić M., Mijić R., Glamočić B., Božović B., Tanasković M., Popović A., Kvalitet životne sredine grada Beograda u 2008. godini (Quality of the Environment in Belgrade in 2008). Secretariat for Environmental Protection of the City of Belgrade, Institute for Public Health and the Regional Environmental Center for Central and Southeast Europe, Belgrade, 2009 (in Serbian with English summery)
Rundić Lj., Knežević S., Banjac N., Ganić M., Milovanović D., Rabrenović D., Geološki objekti i pojave kao integralni deo prirodne i kulturne baštine grada Beograda (Geological objects and phenomena as an integral part of the natural and cultural heritage of the City of Belgrade). Proceedings of the 15th Congress of the Geologists of Serbia, Belgrade, 2010, 711-717 (in Serbian with English summery)
Banjac N., Rundić Lj., Geoturizam – novi vid turistižke ponude na Tari (Geotourism – a new form of tourism on Tara). Geographical Institute “Jovan Cvijić” SASA, Belgrade, 2006 (in Serbian with English summery)
Belij S., Geodiverzitet i geonaslede – savremeni trend razvoja geomorfologije u svetu i kod nas (Geodiversity and geoheritage – the modern trend of development of geomorphology in the world and in our country). Journal of the Geographical Institute “Jovan Cvijic” SASA, 2007, 57, 65-70 (in Serbian)
Marković S.B., Oches E., Sümegi P., Jovanović M., Gaudenyi T., An introduction to the Upper and Middle Pleistocene loess-palaeosol sequences of Ruma section (Vojvodina, Serbia). Quaternary International, 2006, 149, 80-86
Marković S.B., Oches E.A., McCoy W.D., Gaudenyi T., Frechen M., Malacological and sedimentological evidence for “warm” glacial climate from the Irig loess sequence (Vojvodina, Serbia). Geophysics, Geochemistry and Geosystems, 2007, 8, Q09008
Marković S.B., Hambach U., Stevens T., Kukla G.J., Heller F., McCoy W.D., Oches E.A., Buggle B., Zöller L., The last million years recorded at the Stari Slankamen (Northern Serbia) loess-palaeosol sequence: revised chronostratigraphy and long-term environmental trends. Quaternary Science Reviews, 2011, 30(9–10), 1142-1154
Rundić Lj., Geološki objekti i prirodni fenomeni kao integralni elementi geodiverziteta grada Beograda (Geological structures and natural phenomena as integral elements of geological diversity of the City of Belgrade). Faculty of Mining and Geology, Belgrade, 2010 (in Serbian)
Stevanović P., Potok Karagača ispod Avale – klasično mesto nalaska panonske fosilne faune mekušaca (Stream Karagača under Avala Mt. – a classic location of the Pannonian fauna fossil mollusks). Protection of Nature, 1958a, 12, 6-12
Stevanović P., Potok Karagača ispod Avale – klasično mesto nalaska panonske fosilne faune mekušaca (Stream Karagača under Avala Mt. – a classic location of the Pannonian fauna fossil mollusks). Protection of Nature, 1958b, 13, 6-13
Marković S.B., Hambach U., Catto N., Jovanović M., Buggle B., Machalett B., Zoeller L., Glaser B., Frechen M., The Middle and Late Pleistocene loess sequences at Batajnica, Vojvodina, Serbia. Quaternary International, 2009, 198, 1–2, 255-266
Kličković M., Belij S., Petreš D., Trikić M., Simić S., Izveštaj o preliminarnom istraživanju prirodnog jezera Rakina bara u Sremčici kod Beograda (Report on the preliminary study of the natural lake Rakina bara in Sremciča near Belgrade). Institute for nature conservation of Serbia, Belgrade, 2008 (in Serbian with English summery)
Tomić N., Božić S., A modified Geosite Assessment Model (M-GAM) and its Application on the Lazar Canyon area (Serbia). International Journal of Environmental Research, 2014, 8, 4, 1041-1052
Różycka, M., Migoń, P., Customer-Oriented Evaluation of Geoheritage – on the Example of Volcanic Geosites in the West Sudetes, SW Poland. Geoheritage, 2017, 1-15
Vujičić M., Vasiljević B.A., Marković S.B., Hose T.A., Lukić T., Hadzić O. & Janićević S., Preliminary geosite assessment model (GAM) and its application on Fruška Gora Mountain, potential geotourism destination of Serbia. Acta geographica Slovenica, 2011, 51, 2, 361-377
Marković, J. J., The image of Belgrade and Novi Sad as perceived by foreign tourists. Journal of the Geographical Institute “Jovan Cvijić” SASA, 2016, 66, 1, 91-104
Todorović, N., Jovičić D., Motivational factors of youth tourists visiting Belgrade. Journal of the Geographical Institute “Jovan Cvijić” SASA, 2016, 66, 2, 273-289
Census of Population, Households and Dwellings in the Republic of Serbia: Comparative Overview of the Number of Population in 1948, 1953, 1961, 1971, 1981, 1991, 2002 and 2011, Data by settlements” (PDF). Statistical Office of Republic of Serbia, Belgrade. 2011. Retrieved June 27, 2015.
Lukić D., Milovanović D., A contribution to the insight a contribution to the insight into Djerdap geoheritage. In: Cvetković V. (Ed.). Proceedings of the XVI Congress of the Geologists of Serbia, Donji Milanovac, Serbia. Serbian Geological Society, Belgrade, 2014, 877-879
The example of a first part questionnaire (translated in English), which was used for experts’ and visitors’ attitudes toward observed geosites in Belgrade (Serbia). The questionnaire is based on a five point Likert-type scale by rating it from zero to one (0 = not at all important; 0.25 = not very important; 0.50 = neutral; 0.75 = somewhat important; 1.00 = very important). The respondents marked the word in accordance with their attitude towards every presented value.
|Main values (MV)|
|Scientific/Educational value (VSE)||Description||0||0.25||0.50||0.75||1|
|Rarity||Number of closest identical sites.||Common||Regional||National||International||The only occurrence|
|Representativeness||Didactic and exemplary characteristics of the site due to its own quality and general configuration.||None||Low||Moderate||High||Utmost|
|Knowledge on scientific issue||Number of written papers in geo-scientific publications.||None||Local publications||Regional publications||National publications||International publications|
|Level of interpretation||Level of interpretive possibilities on geological and geomorphologic processes, phenomena and shapes and level of scientific knowledge.||None||Moderate level of processes but hard to explain to non-experts),||Good example of processes but hard to explain to non-experts||Moderate level of processes but easy to explain to common visitor||Good example of processes and easy to explain to common visitor|
|Scenic/Aesthetic values (VSA)|
|Viewpoints||Number of viewpoints accessible by a pedestrian pathway. Each must present a particular angle of view and be situated less than 1 km from the site.||Small||-||Medium||-||Large|
|Surface||Whole surface of the site. Each site is considered in quantitative relation to other sites.||-||Low||Medium||High||Utmost|
|Surrounding landscape and nature||Panoramic view quality, presence of water and vegetation, absence of human induced deterioration, vicinity of urban area, etc.||Unfitting||-||Neutral||-||Fitting|
|Environmental fitting||Level of contrast to the nature, contrast of colors, appearance of shapes, etc.||Totally damaged (because of human activities)||Highly damaged (as a result of natural processes)||Medium damaged (with essential geomorphologic features preserved)||Slightly damaged||No damage|
|Current condition||Current state of geosite.||Irreversible (with possibility of total loss)||High (could be easily damaged)||Medium (could be damaged by natural processes or human activities)||Low (could be damaged only by human activities)||None|
|Protection level||Protection by local, regional groups, national government, international organizations, etc.||0||0 to 10||10 to 20||20 to 50||More than 50|
|Vulnerability||Vulnerability level of geosite.||Inaccessible||Low (on foot with special equipment and expert guide tours)||Medium (by bicycle and other means of man-powered transport)||High (by car)||Utmost (by bus)|
|Suitable number of visitors||Proposed number of visitors on the site at the same time, according to surface area, vulnerability and current state.||None||1||2 to 3||4 to 6||More than 6|
The example of a second part questionnaire (translated in English), which was used for experts’ and visitors’ attitudes toward observed geosites in Belgrade (Serbia). The questionnaire is based on a five point Likert-type scale by rating it from zero to one (0 = not at all important; 0.25 = not very important; 0.50 = neutral; 0.75 = somewhat important; 1.00 = very important). The respondents marked the word in accordance with their attitude towards every presented value.
|Additional values (AV) Functional values (VFn)||Description||0||0.25||0.50||0.75||1|
|Accessibility||Possibilities of approaching to the site.||None||Local||Regional||National||International|
|Additional natural values||Number of additional natural values in the in radius of 5 km.||None||Low||Medium||High||Utmost|
|Add. anthropogenic values||Number of additional anthropogenic values in the in radius of 5 km.||None||Local||Regional||National||International|
|Vicinity of emissive centers||Closeness of emissive centers (big urban centers).||None||Less than 12 per year||12 to 24 per year||24 to 48 per year||More than 48 per year|
|Vicinity of road network||Closeness of important road networks in the in radius of 20 km.||More than 50km||50 to 20km||20 to 5km||5 to 1km||Less than 1km|
|Additional functional values||Parking lots, gas stations, mechanics, etc.||None||Low quality||Medium quality||High quality||Utmost quality|
|Touristic values (VTr)|
|Promotion||Level and number of promotional resources.||None||Low||Medium||High||Utmost|
|Organized visits||Annual number of organized visits to the geosite.||None||Low||Medium||High||Utmost|
|Vicinity of visitors’ center||Closeness of visitors’ center to the geosite.||More than 50 km||25-50 km||10-25 km||5-10 km||Less than 5km|
|Interpretative panels||Text and graphics, material quality, size, fitting to surroundings, etc.||More than 25 km||10-25 km||10-5 km||1-5 km||Less than 1 km|
|Number of visitors||Annual number of visitors.||0||0.25||0.5||0.75||1|
|Tourism infrastructure||Pedestrian ways, resting places, garbage cans, toilets, wellsprings. etc.||Common||Regional||National||International||The only occurrence|
|Tour guide service||Expertise level, knowledge of foreign language(s), interpretative skills, etc.||None||Low||Moderate||High||Utmost|
|Hostelry service||Hostelry service close to geosite.||None||Local publications||Regional publications||National publications||International publications|
|Restaurant service||Restaurant service close to geosite.||None||Moderate level of processes but hard to explain to non-experts),||Good example of processes but hard to explain to non-experts||Moderate level of processes but easy to explain to common visitor||Good example of processes and easy to explain to common visitor|